Secondary Logo

Journal Logo

Viewpoint: Remember the ECASS III!

Mosley, Mark MD, MPH

doi: 10.1097/01.EEM.0000464087.58862.2c

Dr. Mosleyis the medical director for residency education at HCA/Wesley Medical Center in Wichita, KS.



The Battle of the Alamo was a sad and rather meaningless loss of human life on both sides, but this defeat of a small army in Texas was used as an emblem and a rallying cry for Texas independence: “Remember the Alamo!” This resounding defeat has been redefined as a victory.

ECASS III is the Alamo of advocates for IV-tPA in stroke. The 3- to 4.5-hour window for IV-tPA in stroke has been repeatedly defeated scientifically, showing increased mortality in the ATLANTIS and IST-3 trials. ECASS III is the only randomized study claiming a “victory” for IV-tPA, yet a more critical look reveals that these data hold the makings of a decisive defeat. The only way to correct the historical record is to continue to discuss it, and the same could be said for the scientific record.

ECASS III was a randomized controlled trial of 821 patients comparing placebo with IV-tPA in the 3- to 4.5-hour window for stroke. The benefit was a 7.2 percent decrease at 90 days compared with placebo using the modified Rankin Scale (mRS) to determine “good” vs “poor” outcome. The harm was “symptomatic” intracerebral hemorrhage (ICH) of 2.4 percent compared with 0.2 percent with placebo. No change in mortality was seen, but the study was underpowered to draw conclusions on mortality. Four major points in ECASS III dramatically alter its authors' conclusions:

Allocation bias: The placebo group in ECASS III was significantly worse at baseline than the intervention group. It had significantly more previous strokes and worse baseline NIH stroke scores. It is no surprise that taking patients with worse brains who have worse strokes are going to have worse outcomes.

Redefining outcomes of good vs poor: A good outcome in the NINDS trial was an mRS score of 0-2; and a poor outcome was an mRS score of 3-6. A good outcome in ECASS III was an mRS of 0-1, and a poor outcome was an mRS of 2-6. If you use the NINDS definitions (which have been the standard) and apply them in the ECASS III trial, no difference in disability is seen between placebo and IV-tPA in spite of a worse baseline with placebo.

Redefining Symptomatic ICH: Using the NINDS trial definition in ECASS III, “symptomatic” ICH is not 2.4 percent but 7.9 percent, which is threefold greater than the ECASS III definition. Regardless of whose definition you use, ICH with IV-tPA in ECASS III was 10 times greater than placebo.

ECASS III Underestimates Bleeding in the Real World: ECASS III excluded all patients on blood thinners regardless of INR and all patients over 80 who do worse with ICH. ECASS III was a carefully controlled trial that has results that are rarely replicated in the real world.

ECASS III data in reality prove to have a negative conclusion for IV-tPA in the 3- to 4.5-hour window (in spite of the authors' conclusion). All randomized controlled trials on thrombolytics for stroke in the timeframe are negative studies, and IV-tPA is currently not FDA-approved for the 3- to 4.5-hour window. The mortality rate in ECASS III was 7.7 percent. The local rates in your own hospital (like ours) are likely to be dramatically higher. There is currently no scientific justification from large randomized controlled trials to offer IV-tPA for stroke in the 3- to 4.5-hour window. What has been waved as a flag of victory for stroke is in reality a sad defeat with meaningless loss of human life. We should all cry (with some irony), “Remember the ECASS III!”

Access the linksin EMN by reading this on our website or in our free iPad app, both available Comments?Write to us

Back to Top | Article Outline


The viewpoint, “ED Are Profitable — Barely,” in last month's issue was written by Laura Medford-Davis, MD, not Cedric Dark, MD. A corrected version of the article is available on our website at EMN regrets the error.

Copyright © 2015 Wolters Kluwer Health, Inc. All rights reserved.